Computational fluid dynamics (CFD) modeling of actual eroded wind turbine blades

Abstract

Abstract. Leading edge erosion (LEE) is one of the most critical degradation mechanisms that occur with wind turbine blades (WTBs), generally starting from the tip section of the blade. A detailed understanding of the LEE process and the impact on aerodynamic performance due to the damaged leading edge (LE) is required to select the most appropriate leading edge protection (LEP) system and optimize blade maintenance. Providing accurate modeling tools is therefore essential. This paper presents a two-part study investigating computational fluid dynamics (CFD) modeling approaches for different orders of magnitudes in erosion damage. The first part details the flow transition modeling for eroded surfaces with roughness on the order of 0.1–0.2 mm, while the second part focuses on a novel study modeling high-resolution scanned LE surfaces from an actual blade with LEE damage on the order of 10–20 mm (approx. 1 % chord); 2D and 3D surface-resolved Reynolds-averaged Navier–Stokes (RANS) CFD models have been applied to investigate wind turbine blade sections in the Reynolds number (Re) range of 3–6 million. From the first part, the calibrated CFD model for modeling flow transition accounting for roughness shows good agreement of the aerodynamic forces for airfoils with leading-edge roughness heights on the order of 140–200 µm while showing poor agreement for smaller roughness heights on the order of 100 µm. Results from the second part of the study indicate that up to a 3.3 % reduction in annual energy production (AEP) can be expected when the LE shape is degraded by 0.8 % of the chord, based on the NREL5MW turbine. The results also suggest that under fully turbulent conditions, the degree of eroded LE shapes studied in this work show the minimal effect on the aerodynamic performances, which results in a negligible difference to AEP.

Transforming Laser-Scanned 750 kW Turbine Surface Geometry Data into Smooth CAD for CFD Simulations

“…Yet another work [7] used 3D-scanning for model aircraft geometry reconstruction, developing an open-source C++ library that, despite its innovation, diverges too significantly from this study for code reuse. Finally, [8] found that erosion-induced irregularities of up to 0.8% chord size at the leading edge may reduce the annual energy production by 3%.…”

Design and implementation of smart integrated hybrid Solar-Darrieus wind turbine system for in-house power generation

“…An unsteady (transient) ''Reynolds-Averaged-Navier-Stokes'' (URANS) method was used as well as a quadratic equation based on a regression model and a rotational degree of freedom solver (6DOF). Obtaining an accurate prediction of the turbine's behavior during startup requires the correct distribution of the lift and drag forces [14,15].…”

An aerodynamic digital twin of real-world leading edge erosion: Acquisition, Generation and 3D CFD

“…Yet the actual topographical manifestation of LE roughness (LER)-with roughness from hereon describing all forms of surface alterations including erosion-needed to assess its aerodynamic impact, is highly probabilistic, as it depends on the interaction of multiple stochastic parameters, like the environmental conditions, material composition and production process. Unfortunately, little high-resolution topographical data of LER from full-scale, operating wind turbines is available [1,4,5,6]. At usual operating Reynolds numbers, at least 50-100 µm spatial resolution is required to capture aerodynamically relevant IOP Publishing doi:10.1088/1742-6596/2767/2/022021 2 surface perturbations [4,7], which can only be provided by high-resolution 3D optical, laser or computed tomography (CT) scanners, making it difficult to obtain such data.…”